管理数据库
create database 等同于 create schema
#导入数据库脚本
MariaDB [db1]> source /root/mysql/hellodb_innodb.sql
管理数据表
数据类型:
1.数字类型
2.时间类型
3.字符串类型
timestamp 只要表中任何一个字段发生修改,自动存储记录时间精确到秒
创建表:
MariaDB [db1]> create table myuser select user,host,password from mysql.user;
表结构一旦定义好,通常不建议进行修改
CASCADE 表示级联删除 一般指的是主外键的关系
选择数据类型
set 多选
enum 单选
复合主键
primary key(name,city)
快速删除(不可恢复)
truncate table students;
生产环境一般用update标志位来代替删除操作
SQL基础语法
查询语句写法不同会造成巨大的性能差别,一条很差的查询可能导致服务器宕机
6记录不是null 7记录才是null null会在对应的位置显示NULL字符 不能通过=null来查询, 只能是 is null 或者 is not null来过滤 MariaDB [db1]> select * from students; +----+------+-------+------+-------+ | id | name | phone | sex | score | +----+------+-------+------+-------+ | 1 | aa | 111 | f | 80 | | 2 | bbb | 2222 | m | 90 | | 3 | cccc | 33333 | f | 80 | | 4 | ffff | 44444 | m | 80 | | 5 | ff | 555 | f | 90 | | 6 | 666 | | m | 40 | | 7 | 777 | NULL | m | 40 |
MariaDB [db1]> select sex from students group by sex; #表示sex这个列有几个不同的值就会被分成几个组 #分完组之后就可以对数据进行相关汇总了 +------+ | sex | +------+ | f | | m | +------+ group by 语法注意点: 1.select 后面只能加分组的列本身 2.各种聚合函数 分组后的过滤条件语法只能使用having 多个分组 MariaDB [db1]> select class,sex,avg(score) from students group by class,sex; +-------+------+------------+ | class | sex | avg(score) | +-------+------+------------+ | 1 | f | 80.0000 | | 1 | m | 85.0000 | | 2 | f | 60.0000 | | 2 | m | 40.0000 | +-------+------+------------+
1.内连接 取两张表的交集 也可以取多张表的交集 select s.name,t.name from students s,teachers t where s.teacherid=t.tid; select s.name,t.name from students s inner join teachers t on s.teacherid=t.tid; 2.交叉连接 两张表的记录分别组合一遍 select * from students cross join teachers; 3.左外连接(谁左谁右十分重要) 默认取左边表的全部记录,取右边表和左边表有交集的记录. 可取左表和右表没有交集的记录 添加一个过滤条件即可 and left.id is null 4.右外连接(谁左谁右十分重要) 取右边表的全部记录,取左边表和右边表有交集的记录. 取右表记录和左表没有交集的记录 and right.id is null 5.自连接 把一张表当成两张表来使用 select s1.name,s2.name from students as s1 inner join students as s2 on s1.id=s2.id 6.union连接 竖直排列组合显示 两个条件: 1.两个select查询的列的数量必须一致 2.两个select查询的对应的列的数据类型必须一致 select id,name from students union select id,name from teachers
select * from students where stuid in(select stuid from scores where score > (select avg(score) from scores));
select查询语句执行流程
1. 先执行from 指定对哪张表进行操作
2. 执行where 过滤表中的行
3 . 执行group by 分组
4. 执行having 对分组后的结果过滤
5. 执行order by 排序
6. 执行select columns 挑选需要显示的字段
7. 执行limit 限定需要显示的记录条数
select查询出来的数据默认是按照磁盘存放次序来排序显示的
select 语法有点类似于awk
= 是精确匹配 where name='abc'
select 可以给字段和表取别名
like 模糊搜索查询
% 任意个任意字符
- 任意单个字符
MariaDB [db1]> select 'hello word' -> ; +------------+ | hello word | +------------+ | hello word | +------------+ 1 row in set (0.00 sec) MariaDB [db1]> select 1+2; +-----+ | 1+2 | +-----+ | 3 | +-----+ MariaDB [db1]> select '1+2',1+2 from user1; +-----+-----+ | 1+2 | 1+2 | +-----+-----+ | 1+2 | 3 | | 1+2 | 3 | | 1+2 | 3 | | 1+2 | 3 | MariaDB [db1]> desc user1; +----------+----------+------+-----+---------+-------+ | Field | Type | Null | Key | Default | Extra | +----------+----------+------+-----+---------+-------+ | user | char(80) | NO | | | | | host | char(60) | NO | | | | | password | char(41) | NO | | | | +----------+----------+------+-----+---------+-------+ MariaDB [db1]> select user as "姓名",host as "主机" from user1; +--------+-----------+ | 姓名 | 主机 | +--------+-----------+ | root | localhost | | root | centos7 | | root | 127.0.0.1 | | root | ::1 | +--------+-----------+ MariaDB [db1]> select class,sex,avg(score) from students group by class,sex; +-------+------+------------+ | class | sex | avg(score) | +-------+------+------------+ | 1 | f | 80.0000 | | 1 | m | 85.0000 | | 2 | f | 60.0000 | | 2 | m | 40.0000 | +-------+------+------------+ MariaDB [hellodb]> select s.name,sc.score,c.course from students s,scores sc,courses c where s.stuid=sc.stuid and sc.courseid=c.courseid limit 5; +-------------+-------+----------------+ | name | score | course | +-------------+-------+----------------+ | Shi Zhongyu | 77 | Kuihua Baodian | | Shi Zhongyu | 93 | Weituo Zhang | | Shi Potian | 47 | Kuihua Baodian | | Shi Potian | 97 | Daiyu Zanghua | | Xie Yanke | 88 | Kuihua Baodian | +-------------+-------+----------------+ MariaDB [hellodb]> select s.name,sc.score,c.course from (select * from students limit 5) s,scores sc,courses c where s.stuid=sc.stuid and sc.courseid=c.courseid; +-------------+-------+----------------+ | name | score | course | +-------------+-------+----------------+ | Shi Zhongyu | 77 | Kuihua Baodian | | Shi Zhongyu | 93 | Weituo Zhang | | Shi Potian | 47 | Kuihua Baodian | | Shi Potian | 97 | Daiyu Zanghua | | Xie Yanke | 88 | Kuihua Baodian | | Xie Yanke | 75 | Weituo Zhang | | Ding Dian | 71 | Daiyu Zanghua | | Ding Dian | 89 | Kuihua Baodian | | Yu Yutong | 39 | Hamo Gong | | Yu Yutong | 63 | Dagou Bangfa | +-------------+-------+----------------+ MariaDB [hellodb]> select re.name,avg(score) from (select s.name,sc.score,c.course from (select * from students limit 8) s,scores sc,courses c where s.stuid=sc.stuid and sc.courseid=c.courseid ) as re group by name order by avg(score) desc; +-------------+------------+ | name | avg(score) | +-------------+------------+ | Shi Qing | 96.0000 | | Shi Zhongyu | 85.0000 | | Xi Ren | 84.5000 | | Xie Yanke | 81.5000 | | Ding Dian | 80.0000 | | Lin Daiyu | 75.0000 | | Shi Potian | 72.0000 | | Yu Yutong | 51.0000 | +-------------+------------+ MariaDB [hellodb]> select c.course,re.stucount from courses as c,(select courseid,count(stuid) as stucount from scores group by courseid) as re where c.courseid=re.courseid; +----------------+----------+ | course | stucount | +----------------+----------+ | Hamo Gong | 3 | | Kuihua Baodian | 4 | | Jinshe Jianfa | 1 | | Taiji Quan | 1 | | Daiyu Zanghua | 2 | | Weituo Zhang | 2 | | Dagou Bangfa | 2 | +----------------+----------+ MariaDB [hellodb]> select * from students where age > (select avg(age) from students); +-------+--------------+-----+--------+---------+-----------+ | StuID | Name | Age | Gender | ClassID | TeacherID | +-------+--------------+-----+--------+---------+-----------+ | 3 | Xie Yanke | 53 | M | 2 | 16 | | 4 | Ding Dian | 32 | M | 4 | 4 | | 6 | Shi Qing | 46 | M | 5 | NULL | | 13 | Tian Boguang | 33 | M | 2 | NULL | | 25 | Sun Dasheng | 100 | M | NULL | NULL | +-------+--------------+-----+--------+---------+-----------+ MariaDB [hellodb]> select name from students where stuid in(select stuid from scores where courseid in(1,2,4,7)); +-------------+ | name | +-------------+ | Shi Zhongyu | | Shi Potian | | Xie Yanke | | Ding Dian | | Yu Yutong | | Shi Qing | | Xi Ren | | Lin Daiyu | +-------------+ MariaDB [hellodb]> select s.name,s.age from students s,(select classid,avg(age) as age,count(stuid)from students group by classid having count(stuid)>=3) as re where s.classid = re.classid and s.age > re.age; +---------------+-----+ | name | age | +---------------+-----+ | Shi Potian | 22 | | Xie Yanke | 53 | | Ding Dian | 32 | | Yu Yutong | 26 | | Yuan Chengzhi | 23 | | Xu Zhu | 21 | | Lin Chong | 25 | | Hua Rong | 23 | | Huang Yueying | 22 | +---------------+-----+ 9 rows in set (0.00 sec)
视图
VIEW,虚表 保存有实表的查询结果
视图中的数据事实上存储于“基表”中 因此,其修改操作也会针对基表实现; 其修改操作受基表限制
函数
系统函数和自定义函数 保存在mysql.proc表中
存储过程
存储过程 存储过程保存在mysql.proc表中
触发器
trigger_name：触发器的名称
trigger_time：{ BEFORE | AFTER },表示在事件之前或之后触发
trigger_event:：{ INSERT |UPDATE | DELETE },触发的具体事件
tbl_name：该触发器作用在表名
触发器的执行不是由程序调用, 也不是由手工启动,而是由事件来触发,激活从而实现执行
触发器实例,在向学生表INSERT数据时,学生数增加,DELETE学生时,学生数减少
MySQL用户和授权管理
用户管理
创建用户：CREATE USER
CREATE USER 'USERNAME'@'HOST' [IDENTIFIED BY 'password']； 默认权限：USAGE 只能连接不能进行任何操作
用户重命名：RENAME USER RENAME USER old_user_name TO new_user_name
删除用户： DROP USER 'USERNAME'@'HOST‘
示例：删除默认的空用户 DROP USER ''@'localhost';
修改密码：
mysql>SET PASSWORD FOR 'user'@'host' = PASSWORD(‘password');
mysql>UPDATE mysql.user SET password=PASSWORD('your_password') WHERE clause;
此方法需要执行下面指令才能生效：mysql> FLUSH PRIVILEGES;
#mysqladmin -u root –poldpass password ‘newpass‘
忘记管理员密码的解决办法：
启动mysqld进程时,为其使用如下选项: --skip-grant-tables --skip-networking 修改配置文件/etc/my.cnf中的mysqld
使用UPDATE命令修改管理员密码
关闭mysqld进程,移除上述两个选项,重启mysqld
权限管理
授权和创建用户可以都通过grant命令一步来实现,因为grant授权的用户如果不存在,mysql会自动创建对应的用户
权限类别：
1.数据库级别
2.表级别
3.字段级别
4.管理类
5.程序类
授权语法:
GRANT priv_type [(column_list)],... ON [object_type] priv_level TO 'user'@'host' [IDENTIFIED BY 'password']
priv_type: ALL [PRIVILEGES]
object_type:TABLE | FUNCTION | PROCEDURE
priv_level: *(所有库) | *.* | db_name.* | db_name.tbl_name | tbl_name(当前库的表) | db_name.routine_name(指定库的函数,存储过程,触发器)
GRANT SELECT (col1), INSERT (col1,col2) ON mydb.mytbl TO 'someuser'@'somehost';
MariaDB [(none)]> grant all on hellodb.* to yxh@"192.168.%.%" identified by 'root';
MariaDB [(none)]> grant select,insert on hellodb.* to laoge@'%' identified by 'root';
MariaDB [(none)]> grant select(stuid,name) on hellodb.students to dage@'%' identified by 'root';
MariaDB [(none)]> show grants for yxh@"192.168.%.%";
回收授权语法:
REVOKE priv_type [(column_list)] [, priv_type [(column_list)]] ... ON [object_type] priv_level FROM user [, user] ...
示例: REVOKE DELETE ON testdb.* FROM 'testuser'@'%'
查看用户的授权信息:
Help SHOW GRANTS
SHOW GRANTS FOR 'user'@'host';
SHOW GRANTS FOR CURRENT_USER[()];
注意：MariaDB服务进程启动时会读取mysql库中所有授权表至内存
(1) GRANT或REVOKE等执行权限操作会保存于系统表中,MariaDB的服务进程通常会自动重读授权表,使之生效
(2) 对于不能够或不能及时重读授权表的命令,可手动让MariaDB的服务进程重读授权表
mysql> FLUSH PRIVILEGES;